公开(公告)号：US09754036B1

公开(公告)日：2017-09-05

申请号：US14226027

申请日：2014-03-26

Applicant: Google Inc.

Inventor： Evgeniy Gabrilovich ,  John Giannandrea ,  Jason Douglas

IPC: G06F17/30 ,  G06Q30/02

CPC classification number: G06F17/30867 ,  G06Q30/0256

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for a search engine to provide an entity some indication of topics in which a user may have an interest. The methods, systems, and apparatus include actions of receiving information at a search engine from a classifier indicating that a user is likely interested in a set of topics based on information about the user provided by the search engine to the classifier. Additional actions may include selecting a subset of the topics and generating a search results page that includes one or more references for one or more resources that are responsive to a search query. At least one reference of the one or more references may include information based on the received information that indicates that the user is likely interested in the subset of topics.

公开(公告)号：US09728185B2

公开(公告)日：2017-08-08

申请号：US14720113

申请日：2015-05-22

Applicant: Google Inc.

Inventor： Johan Schalkwyk ,  Francoise Beaufays ,  Hasim Sak ,  John Giannandrea

IPC: G10L15/16 ,  G10L15/18 ,  G10L15/26 ,  G06N3/02 ,  G10L15/187

CPC classification number: G10L15/18 ,  G06N3/02 ,  G06N3/0445 ,  G10L15/16 ,  G10L15/187 ,  G10L15/26

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for recognizing speech using neural networks. One of the methods includes receiving an audio input; processing the audio input using an acoustic model to generate a respective phoneme score for each of a plurality of phoneme labels; processing one or more of the phoneme scores using an inverse pronunciation model to generate a respective grapheme score for each of a plurality of grapheme labels; and processing one or more of the grapheme scores using a language model to generate a respective text label score for each of a plurality of text labels.

公开(公告)号：US20150340034A1

公开(公告)日：2015-11-26

申请号：US14720113

申请日：2015-05-22

Applicant: Google Inc.

Inventor： Johan Schalkwyk ,  Francoise Beaufays ,  Hasim Sak ,  John Giannandrea

IPC: G10L15/18 ,  G10L15/26

CPC classification number: G10L15/18 ,  G06N3/02 ,  G06N3/0445 ,  G10L15/16 ,  G10L15/187 ,  G10L15/26

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for recognizing speech using neural networks. One of the methods includes receiving an audio input; processing the audio input using an acoustic model to generate a respective phoneme score for each of a plurality of phoneme labels; processing one or more of the phoneme scores using an inverse pronunciation model to generate a respective grapheme score for each of a plurality of grapheme labels; and processing one or more of the grapheme scores using a language model to generate a respective text label score for each of a plurality of text labels.

Abstract translation: 方法，系统和装置，包括在计算机存储介质上编码的计算机程序，用于使用神经网络识别语音。 其中一种方法包括接收音频输入; 使用声学模型处理音频输入以为多个音素标签中的每一个产生相应的音素分数; 使用反向发音模型处理一个或多个音素得分，以产生多个图形标签中的每一个的各自的图形分数; 以及使用语言模型处理一个或多个所述图形分数，以生成多个文本标签中的每一个的相应文本标签分数。